Neuron Explorer

Qwen3-0.6B

the word 'number' after 'the'

A neuron in layer 22 of Qwen3-0.6B. Its strongest snippets are very consistent: they mostly peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron responds most to the word 'number' right after 'the', as in 'the number of samples' or 'the number of correct answers'. Turning it up produced rambling text that did not show a clear link to that word.

Written by Claude from this neuron's activations and steering outputs. Confidence: medium. Think it is wrong? Flag it (opens a prefilled issue on GitHub).

Fires most on

The text windows where this neuron's activation was highest. Amber marks positive activation; the outlined token is the peak. Hover a token for its strength.

  1. #1 peak

    of 0.02). Plug your population size and required margin of error into the formula. The result will be the number of samples you need to take.⏎For example, suppose that you have a group

  2. #2 peak

    help judge the effectiveness of the course, Jessica measured the number of words per minute she could read prior to enrolling in the course. She obtained the following five number summary:⏎Use this information to draw a boxplot of the reading speed.⏎Answer

  3. #3 peak

    ’s First⏎Before you learn how to calculate and monitor your target training heart rate, you have to know your resting heart rate. Your resting heart rate is the number of times your heart beats per minute while

  4. #4 peak

    , mathematical textbooks typically refer to nodes as vertices. They are also more precise in mathematical notation. For example, this chapter will use the symbol N to denote the number of nodes in the graph. In a mathematics context, the set of nodes is

  5. #5 peak

    number is to find the value of the mark. You can do this by using a percentage calculator. You can

  6. #6 peak

    f(), and ldexpl() return ±HUGE_VAL, ±HUGE_VALF, and ±HUGE_VALL (according to

Steering

Push this neuron up or down while the model writes, and watch what changes. The outputs were generated ahead of time, so nothing runs in your browser.

turned down a lot
Prompt

Here is something I have been thinking about lately: the power word "power" — the power word power word power word power word power word power word — the word power word word power word word power word power word power word word power word word word power word word word word power word word

Most negative on

Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.

  1. #1 peak

    Information:||Web: http://www.tandfonline.com/pricing/journal/VECE20|⏎When requesting a correction, please mention this item's handle: RePEc:taf:jeduce:v:42:y

  2. #2 peak

    that the groups with the most weight loss were the diet-only and diet and exercise group with women in those groups losing an average of 10 percent of their starting weight, according to the study.⏎Additionally, the women

  3. #3 peak

    directly to that person rather than through a companion or interpreter. Conduct interviews in a manner that emphasizes abilities, achievements and individual qualities.⏎Address people who have disabilities by their first names only when calling everyone present by

  4. #4 peak

    -2 viruses were diverse and changed over time, but that the percentage of infected participants that was<|endoftext|>If you are participating with a teacher, please enter your teacher's code here:⏎An email has been sent to you (or your parent if you

Connections

Show on map

Which neurons in earlier layers feed this one, and which later ones it feeds. Amber links excite, blue links suppress; longer bars are stronger.

Fed by earlier layers

  1. L19:603 suppresses
  2. L19:2761 suppresses
  3. L18:2718 excites
  4. L19:1176 excites
  5. L19:1928 excites
  6. L20:996 excites
  7. L18:2765 excites
  8. L19:870 excites
  9. L18:1304 suppresses
  10. L20:1284 suppresses
  11. L18:1555 excites
  12. L19:1378 suppresses
  13. L19:862 excites
  14. L20:706 suppresses
  15. L18:2049 excites
  16. L16:1676 excites
  17. L16:2258 suppresses
  18. L19:1514 excites
  19. L20:2314 suppresses
  20. L14:258 excites
  21. L17:1979 excites
  22. L19:1233 excites
  23. L21:2555 excites
  24. L21:449 suppresses

Feeds later layers

  1. L27:2084 suppresses
  2. L23:94 excites
  3. L26:2244 suppresses
  4. L27:612 weakly suppresses
  5. L24:78 weakly suppresses
  6. L27:425 weakly suppresses
  7. L25:417 weakly suppresses
  8. L24:574 weakly suppresses
  9. L25:1651 weakly suppresses
  10. L24:1374 weakly suppresses
  11. L23:926 weakly suppresses
  12. L23:134 weakly suppresses
  13. L27:1143 weakly suppresses
  14. L23:2778 weakly suppresses
  15. L25:1391 weakly suppresses
  16. L24:418 weakly suppresses
  17. L25:2922 weakly suppresses
  18. L23:817 weakly suppresses
  19. L27:1752 weakly suppresses
  20. L25:1114 weakly suppresses
  21. L24:121 weakly excites
  22. L26:507 weakly suppresses
  23. L27:393 weakly suppresses
  24. L25:210 weakly suppresses

Activation histogram

How often this neuron fires at each strength. Almost always it does close to nothing (the tall bars in the middle); the snippets above are the rare moments it fired hard, amber for positive and blue for negative.

1 values between -18.30 and -17.551 values between -16.80 and -16.061 values between -13.81 and -13.071 values between -13.07 and -12.322 values between -9.33 and -8.581 values between -8.58 and -7.843 values between -7.84 and -7.093 values between -7.09 and -6.341 values between -6.34 and -5.594 values between -5.59 and -4.856 values between -4.85 and -4.103 values between -4.10 and -3.3510 values between -3.35 and -2.6023 values between -2.60 and -1.8644 values between -1.86 and -1.11200 values between -1.11 and -0.362847 values between -0.36 and 0.39707 values between 0.39 and 1.1327 values between 1.13 and 1.8810 values between 1.88 and 2.634 values between 2.63 and 3.377 values between 3.37 and 4.121 values between 4.87 and 5.621 values between 10.85 and 11.59offnegativepositiveoften

Logit lens

The words this neuron nudges the model toward (amber) or away from (blue) when it fires. A hint, not a label: many neurons work through later layers instead.

Promotes

  1. dating
  2. cmds
  3. otherwise
  4. of
  5. пут
  6. gratuites
  7. Cars
  8. OfString
  9. śmy
  10. abra

Suppresses

  1. ilar
  2. autical
  3. ycl
  4. 院
  5. ctal
  6. pirit
  7. :relative
  8. mlink
  9. sembled
  10. nces